Voice Recognition System for Healthcare with Weighted Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice recognition systems in healthcare settings face inaccuracies due to pronunciation errors and fast speech, requiring manual correction, which is inefficient and labor-intensive.
Innovation Solution
A voice recognition system that includes a processor and storage device with modules for identification, comparison, scoring, and engine training, which adjusts word error rates based on important word weights to calculate a professional score, automatically identifying and weighting important keywords for improved accuracy and automated training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If voice recognition is used to input text through speaking, then recording efficiency is improved, but recognition accuracy deteriorates due to pronunciation errors and fast speech
Solution Approach 1:
The system calculates a word error rate by comparing recognized text with correct results, then uses this feedback to identify important words and adjust recognition weights. This closed-loop feedback mechanism continuously improves recognition accuracy while maintaining efficient voice input.
Solution Approach 2:
The system dynamically adjusts recognition parameters by assigning different weights to different words based on their importance. By changing the weight parameters of specific words in the recognition model, the system optimizes accuracy for critical terms while maintaining overall recognition efficiency.
2Measurement precision
If manual modification is required to correct recognition errors, then recognition accuracy is improved, but time consumption and labor costs increase
Solution Approach 1:
The system automatically identifies important words and adjusts recognition weights without requiring manual intervention. The automated importance word identification and weight adjustment mechanisms enable the system to self-optimize recognition accuracy, eliminating the need for time-consuming manual corrections.
Solution Approach 2:
The system replaces manual correction processes with automated computational methods. By using algorithms to identify important words and adjust recognition parameters, the system substitutes mechanical human labor with automated processing, significantly reducing time consumption while maintaining or improving accuracy.
3Reliability
If automated training is implemented using high-scoring corpora, then recognition capability is improved, but system complexity increases
Solution Approach 1:
The system pre-processes and scores corpora to identify high-quality training data before actual training occurs. By performing preliminary scoring and selection of important words and high-scoring corpora, the system prepares optimized training datasets in advance, making the subsequent training process more efficient and manageable despite increased functionality.
Data Source
AI summary
A voice recognition method includes the following steps. An audio and a correct result are received. The audio is recognized, and a text file corresponding to the audio is output. The word error rate is determined by comparing the text file to the correct result. The word error rate is adjusted according to the weight of at least one important word, in order to calculate a professional score that corresponds to the text file. A determination is made as to whether the professional score is higher than a score threshold. In response to the professional score is higher than the score threshold, the text file, the audio, or the correct result corresponding to the professional score is sent to an engine training module for training.